Demand response can include voluntary changes in electricity usage of an electric utility customer (e.g., residential and/or commercial structure) to better match the demand for power with the supply. Demand response can allow for the reduction of load on electrical grids based on signals that are sent to structures. For example, a demand response request (e.g., signal) sent to a structure by an electricity supplier or other grid management entity (e.g., utility) can result in some number of loads in that facility being shut down and/or running at reduced capacity.
Previous approaches to demand response have shortcomings in a number of respects. In order to effectively manage demand response, utilities may validate impacts of demand response events in terms of overall energy shed. The shed can be calculated as the difference between aggregated load under the demand response signal and a baseline load (e.g., the aggregated load if no demand response signal was sent). Previous approaches to determining the baseline may exhibit high error rates due to their usage of consumption data from non-demand response days, which may experience different weather and/or energy usage than demand response days.
Additionally, previous approaches may not assess which demand response program(s) are appropriate for a given customer. Stated in other terms, previous approaches may not assess which demand response program(s) the utility should offer and/or recommend to the customer because they are likely to bring the most value to the utility.
Additionally, previous approaches may not leverage thermostat data in order to identify customers who may benefit the most from an energy audit. Some previous approaches select candidate customers for an energy audit using information including customer address, name, age, and/or electricity bills, and then, upon approval by the customer, send an auditor to the structure having little or no context as to the reason(s) for the inefficiencies. Increased time spent at the structure translates to increased costs.